Quantum computer calculates exact energy of molecular hydrogen

I've heard this dozens of times, and I know every time that it's crazy.

But I haven't really managed to figure out what part of this 'rumor' is actually true.

Do you mind explaining where this idea comes from, and whether it's remotely right, or what?

What do you mean?

Quantum mechanics is indeed a different viewpoint

As I recall, the Feynman diagram approach assumes that every conceivable interaction between particles that could happen does happen with a given probability depending on the interaction's complexity.

Indeed, the Heisenberg Uncertainty Principle is quite a mind fuck - if you know the exact speed of a particle, the location of particle become utterly uncertain to the point that it could be anywhere in the universe.

So what specifically is your problem with the above quote?

"This has got to stop."

Popular science journalists always fall back on the simplified multi-universe explanation to explain quantum computing. I wish they would adopt another paradigm too, but in general I want to see a better quality of synopses of technical papers in popular journalism.

Getting back to the subject of the article... Scott Aaronson has always said the most interesting use of quantum computing would be to simulate natural processes. This may be the first significant achievement of that. (At least that I have heard of.)

I doubt it will ever stop. It's too appealing for journalists, and a lot easier than explaining the true state of affairs. I have story about this which I can't resist telling. It really annoyed me at the time, but with a few years distance I think it's both funny and illustrative of media stories about quantum computing.

In 2001, I was interviewed about quantum computing by a reporter (from the Sydney Morning Herald, if I recall correctly) who repeatedly asked me for a "simple, concrete explanation" of how quantum computers work.

If you take this request really (really!) literally, such an explanation probably isn't possible. If it was, you could efficiently simulate the elements of the explanation on a conventional computer, and quantum computers would offer no real advantage over conventional computers. In fact, in my opinion, what's interesting about quantum computing is precisely the gap between our ability to give such a simple, concrete explanation and what's really going on. You _can_ give a (fairly) simple indirect explanation of how they work - simple if someone's comfortable with complex numbers and vectors - but it is pretty indirect.

I explained this to the reporter, who made all the right noises, and then gave him the "simultaneously in all possible states" explanation as an example of the kind of nonsense that sounds good, but actually obscures the truth of what's going on.

You can probably guess the rest: when the article came out, I was in there, quoted saying something along the lines of "When a quantum computer is put to work on a problem, it considers all possible answer by simultaneously arranging..."

The problem is that this line is just too tempting and pat a story for reporters. It creates the illusion of understanding (in my opinion), while leaving the more complex truth undisturbed.

If it scale and lives up to the general description it will be. Whether it will do that it another question.

How can I read the full article?

I don't have the time right now to slog through what is probably a very interesting paper, but this did catch my eye "We perform a key algorithmic step the iterative phase estimation algorithm in full, achieving a high level of precision and robustness to error. We implement other algorithmic steps with assistance from a classical computer and explain how this non-scalable approach could be avoided."

It sounds like they claim at least part of it they did with real quantum computing. Scanning the paper I think they did some calculations on "real" one and two qubit systems.